Pain has been characterized and described in various different ways in the literature. For example, pain can be intense, localized, sharp or stinging, and/or dull, aching, diffuse or burning in nature. Pain can also be centralized, taking place in the dorsal horn of the spinal cord, the brain stem and brain, or peripheral, taking place at the injury site and surrounding tissue. Pain that occurs for extended periods of time (i.e., persistent) is generally referred to as chronic pain. Examples of chronic pain include neuropathic pain, inflammatory pain, and cancer pain. These pains can be related to hyperalgesia and/or allodynia, where hyperalgesia refers to an increase in sensitivity to a typically noxious stimulus and allodynia refers to an increase in sensitivity to a typically non-noxious stimulus.
A type of chronic pain that currently lacks adequate pharmacological treatment is neuropathic pain. Neuropathic pain is generally thought of as a chronic pain caused by damage to or pathological changes in the peripheral or central nervous systems. Examples of pathological changes related to neuropathic pain include prolonged peripheral or central neuronal sensitization, central sensitization related damage to nervous system inhibitory and/or excitatory functions and abnormal interactions between the parasympathetic and sympathetic nervous systems. A wide range of clinical conditions may be associated with or form the basis for neuropathic pain including for example diabetes, post traumatic pain of amputation, lower back pain, cancer, chemical injury or toxins, other major surgeries, peripheral nerve damage due to traumatic injury compression, nutritional deficiencies, or infections such as shingles or HIV.
There are various types of agents currently being used to treat pain such as for example, non-narcotic analgesics such as aspirin, acetaminophen or ibuprofen; non-steroidal anti-inflammatory drugs (NSAID); narcotic analgesics such as morphine, hydromorphone, fentanyl, codeine or meperidine; steroids such as prednisone or dexamethasone; tricyclic antidepressants such as amitriptyline, desipramine, or imipramine; antiepileptics such as gabapentin, carbamazepine, topiramate, sodium valproate or phenytoin; or combinations of these different agents. However, these agents are typically unsatisfactory for treating pain of a chronic nature, and can have adverse effects such as drowsiness, dizziness, dry mouth, weight gain, memory impairment, and/or orthostatic hypotension.
Of more recent interest has been the use of inhibitors of the N-methyl-D-aspartate (“NMDA”) receptors to treat pain (hereinafter called “NMDA receptor antagonists”). It has been shown that NMDA receptors are involved in a wide range of processes including, neuronal death following ischemia, synaptic plasticity associated with memory formation and central sensitization during persistent pain. It is believed that glutamate, which regulates NMDA receptors, plays a key role in pain and especially chronic pain.
NMDA receptors are localized throughout the central nervous system. NMDA receptors are ligand-gated cation channels that modulate sodium, potassium and calcium ions flux when they are activated by glutamate in combination with glycine. Structurally, the NMDA receptor is thought to be comprised of heteromultimeric channels containing two major subunits designated as NR1 and NR2. These subunits contain a glycine binding site, a glutamate binding site and polyamine binding site. For the NR1 subunit, multiple splice variants have been identified, whereas for the NR2 subunit, four individual subunit types (NR2A, NR2B, NR2C, and NR2D) have been identified. The NMDA receptor also contains a Mg++ binding site located inside the pore of the ionophore of the NMDA receptor/channel complex, which blocks the flow of ions. Phencyclidine, as well as other compounds, appear to bind to this Mg++ site. In order for PCP to gain access to the PCP receptor, the channel must first be opened by glutamate and glycine (i.e., use dependence).
Various NMDA antagonists have been developed to interact with these sites of the NMDA receptor. For example, NMDA receptor glutamate site antagonists refer to those antagonists that interact with the glutamate binding site of the NR2 subunit. Examples of NMDA receptor glutamate site antagonists that have been shown in preclinical models to suppress pain include CGS-19755 (Selfotel; cis-4-phosphonomethyl-2-piperidine carboxylic acid), CPP (3-(2-carboxypiperazinyl-4-yl)propyl-1-phosphonic acid) and AP5 (D-2 amino 5-phosphonopentanoic acid). See e.g., Karlsten and Gordh, Drugs and Aging 11: 398–412, (1997). Other NMDA receptor antagonists have been identified that interact at the strychinine-in-sensitive glycine site (glycineβ) such as L701324 (7-chloro-4-hydroxy-3-(3-phenoxy)phenyl-2(1H)-quinoline) and at the polyamine site such as ifenprodil. Noncompetitve NMDA receptor channel blocking antagonists that have been found effective in suppressing pain include dextromethorphan, ketamine, memantine, and amantadine. See e.g., Hao et al., Pain 66:279–285 (1996); Chaplan et al., J. Pharmacol. Exper. Ther. 280:829–838 (1997); Suzuki et al., Pain 91:101–109, (2000); Bennett, J. Pain Symptom Management 19: S2 (2000); Sang, J. Pain Symp. Manag. 19 (1): S21, (2000).
NMDA receptor antagonists have been used in clinical settings to treat pain. For example, ketamine has been used to treat postherpetic neuralgia pain, phantom limb pain, post nerve injury pain, postoperative pain, and burn pain. Also, for example dextromethorphan has been used to treat diabetic neuropathy pain, and postoperative pain; and amantadine has been used to treat pain in cancer patients.
Clinical usefulness of these NMDA receptor antagonists have been limited by adverse effects such as headache, disturbances of motor function such as ataxia, sedation and/or psychotomimetic effects such as dizziness, hallucinations, dysphoria, or disturbances of cognitive function at analgesic doses. See e.g., Hao et al., Pain 66:279–285 (1996); Chaplan et al., J. Pharmacol. Exper. Ther. 280:829–838 (1997); Suzuki et al., Pain 91:101–109, (2000); Bennett, J. Pain Symptom Management 19: S2 (2000); Sang, J. Pain Symp. Manag. 19 (1): S21, (2000). For example, the high affinity NMDA receptor channel blocker ketamine, which is occasionally used for burn related pain has reported adverse effects that has limited its use in patients (Pal et al., Burns 23: 404–412, 1997). Additionally, development of the NMDA receptor channel blocking antagonist dizocilpine (MK-801) was terminated because of psychotomimetic effects similar to those produced by phencyclidine (i.e., PCP). It has been suggested that lower affinity channel blockers such as dextromethorphan, amantadine and memantine might have fewer adverse effects than the high affinity blockers (Rogawski, Trends Pharmacol. Sci. 14:325, 1998). In support of this view, dextromethorphan had analgesic effects in patients suffering from diabetic neuropathy with fewer side effects than ketamine (Sang, J. Pain Symp. Manag. 19 (1): S21, 2000). Similarly, amantadine relieved surgical neuropathic pain in cancer patients with fewer side effects (Hewitt, Clin. J. Pain 16: 573, 2000).
However, even with the lower affinity noncompetitve NMDA receptor channel blocking antagonists, like the higher affinity noncompetitive antagonists, there have been undesirable psychotomimetic effects which have hampered development. For example, in preclinical models, NMDA receptor channel blockers of varying affinities consistently produce PCP-like discriminative stimulus effects in rats trained to discriminate between saline and PCP. Memantine, ketamine and dizocilpine all substitute for the PCP-like discriminative stimulus effects in rats (Nicholson et al., Behav. Pharmacol. 9(3): 231–243, 1998; Mori et al., Behav. Brain Res. 119:33–40, 2001). Moreover, like PCP, memantine maintains self-administration in monkeys suggesting that it might have abuse potential in humans (Nicholson et al., Behav. Pharmacol. 9(3): 231–243, 1998). Use dependent NMDA receptor channel blockers can also increase heart rate and blood pressure, which can further limit their clinical utility.
Although NMDA receptor glutamate antagonists do not have the degree of psychotomimetic side effects in humans or PCP-like discriminative stimulus effects in non-humans (see e.g., Baron and Woods, Psychopharmacol. 118(1): 42–51, (1995); Mori et al., Behav. Brain Res. 119:33–40, (2001); France et al., J. Pharmacol. Exper. Ther. 257(2): 727–734, (1991); France et al., Eur. J. Pharmacol. 159(2): 133–139, (1989)), they have been shown to have many undesirable side effects. For example, the NMDA glutamate antagonist CGS-19755 has been shown to have a transient, reversible induction of vacuoles in some layers of the cingulate and retrosplenial cortices of mice and rats at behaviorly effective doses (i.e., effectiveness/vacuolization ratio of 1). See e.g., Herring et al., Excitatory Amino Acids Clinical Results with Antagonists, published by Academic Press, Chptr 1 (1997). Although the functional implications of vacuolization are unclear, previous studies suggest that this vacuolization correlates with the psychotomimetic effects produced by NMDA receptor antagonists (see e.g., Olney et al., Science, 244: 1630–1632, 1989; Olney et al., Science 254: 1515–1518, 1991) and might lead to limited neuronal cell death as in the case of dizocilpine (Fix et al., Exp. Neurol. 123: 204–215, 1993).
Thus, it would be desirable to find alternative compounds effective in treating pain. Preferably these compounds would have reduced adverse side effects and/or be more effective in treating pain.
U.S. Pat. No. 5,168,103 to Kinney et al. (hereinafter “Kinney”) discloses certain [[2-(Amino-3,4-dioxo-1-cyclobuten-1-yl)amino]alkyl]-acid derivatives useful as neuroprotectant and anticonvulsant agents. These [[2-(Amino-3,4-dioxo-1-cyclobuten-1-yl)amino]alkyl]-acid derivatives are disclosed as competitive NMDA antagonists useful to treat certain central nervous system disorders such as convulsions, brain cell damage and related neurodegenerative disorders.
Side effects of one of the compounds disclosed in the Kinney patent, [2-(8,9-dioxo-2,6-diazabicyclo[5.2.0]non-1(7-en-2-yl)ethyl]phosphonic acid, were previously evaluated in healthy volunteers in a phase I study conducted in Europe. This study was in connection with developing this compound for treating stroke-related ischemia in patients (Bradford et al., Stroke and Cerebral Circulation abstract, 1998).
The present inventors have found that the cyclobutene derivatives in Kinney are effective in treating pain in a variety of preclinical pain models. For example, the present inventors have found that these cyclobutene derivatives can relieve pain under conditions where comparitor NMDA receptor antagonists tested herein do not. Additionally, these cyclobutene derivatives surprisingly do not have the degree of adverse side effects exhibited by known NMDA receptor antagonists at dosages needed for pain relief.
For example, the present inventors, as described in more detail hereinafter, have found that compounds disclosed in Kinney, such as [2-(8,9-dioxo-2,6-diazabicyclo[5.2.0]non-1(7-en-2-yl)ethyl]phosphonic acid, did not produce ataxia or sedation in comparison to other reported competitive glutamate antagonists (CGS-19755), competitive polyamine antagonists (ifenprodil) and use dependent channel blockers (MK-801, memantine; dizocilipine, ketamine) at doses needed to relieve pain in preclinical models. Additionally, as mentioned previously, some NMDA receptor antagonists, such as CGS-19755 were found to exhibit a transient, reversible induction of vacuoles in some layers of the cingulate and retrosplenial cortices of mice and rats. In contrast to CGS-19755, which caused vacuolization at behaviorally effective doses, cyclobutene derivatives such as [2-(8,9-dioxo-2,6-diazabicyclo[5.2.0]non-1(7-en-2-yl)ethyl]phosphonic acid had an effectiveness/vacuolization ratio as large as 16. Moreover, unlike the NMDA receptor channel blocking antagonists previously mentioned herein, the cyclobutene derivatives such as [2-(8,9-dioxo-2,6-diazabicyclo[5.2.0]non-1(7-en-2-yl)ethyl]phosphonic acid did not substitute for PCP in rats, suggesting that this compound would not be associated with PCP-like psychotomimetic effects or contain PCP-like abuse liability. Additionally, [2-(8,9-dioxo-2,6-diazabicyclo[5.2.0]non-1(7-en-2-yl)ethyl]phosphonic acid was devoid of many PCP-like effects up to doses 4–10 times higher than those effective in an ischemia model.